src/scoring_engine.py

import json
import logging
from datetime import datetime
from collections import defaultdict

logger = logging.getLogger(__name__)

class ContextualScoringEngine:
    """
    Moteur de scoring qui maintient la compatibilité avec l'ancienne interface
    tout en offrant les nouvelles fonctionnalités.
    """

    ALPHA = 0.5  
    BETA = 0.3    
    GAMMA = 0.2  

    CONTEXT_VALUES = {
        "formations": 0.3,
        "projets": 0.6,
        "experiences_professionnelles": 0.8,
    }

    def __init__(self, parsed_cv_data: dict):
        self.cv_data = parsed_cv_data.get("candidat", {})
        if not self.cv_data:
            raise ValueError("Données du candidat non trouvées dans le CV parsé.")

    def _normalize_score(self, value: float) -> float:
        """Normalise une valeur sur une échelle de 0 à 1."""
        return 1 - (1 / (1 + float(value)))

    def _parse_date(self, date_str: str) -> datetime | None:
        """Parse une date de manière robuste."""
        if not date_str or not isinstance(date_str, str):
            return None
            
        date_str_lower = date_str.lower()
        if date_str_lower in ["aujourd'hui", "maintenant", "en cours", "current"]:
            return datetime.now()
        
        for fmt in ("%m/%Y", "%Y"):
            try:
                return datetime.strptime(date_str, fmt)
            except ValueError:
                continue
        return None

    def _calculate_duration_in_years(self, start_date_str: str, end_date_str: str) -> float:
        """Calcule la durée d'une expérience en années."""
        start_date = self._parse_date(start_date_str)
        end_date = self._parse_date(end_date_str)
        
        if start_date and end_date:
            if end_date < start_date:
                return 0.0
            return (end_date - start_date).days / 365.25
        return 0.0

    def calculate_scores(self) -> dict:
        """
        Calcule les scores pour toutes les compétences.
        Maintient la compatibilité avec l'ancienne interface.
        """
        skills_data = self.cv_data.get("compétences", {})
        skills_list = []
        
        if isinstance(skills_data, dict):
            skills_list.extend(skills_data.get("hard_skills", []))
            skills_list.extend(skills_data.get("soft_skills", []))
        elif isinstance(skills_data, list):
            skills_list = [item.get("nom") for item in skills_data if item.get("nom")]
        
        if not skills_list:
            logger.warning("Aucune compétence à analyser dans le CV.")
            return {"analyse_competences": []}

        skill_metrics = {
            skill.lower(): {
                "original_name": skill,
                "contexts": set(),
                "frequency": 0,
                "max_duration": 0.0
            }
            for skill in skills_list if skill
        }

        experiences_key = "expériences" if "expériences" in self.cv_data else "experiences_professionnelles"
        for exp in self.cv_data.get(experiences_key, []):
            exp_text = json.dumps(exp, ensure_ascii=False).lower()
            duration = self._calculate_duration_in_years(
                exp.get("date_debut", exp.get("start_date", "")), 
                exp.get("date_fin", exp.get("end_date", ""))
            )
            
            for skill in skill_metrics:
                if skill in exp_text:
                    skill_metrics[skill]["contexts"].add("experiences_professionnelles")
                    skill_metrics[skill]["frequency"] += exp_text.count(skill)
                    if duration > skill_metrics[skill]["max_duration"]:
                        skill_metrics[skill]["max_duration"] = duration

        projects_data = self.cv_data.get("projets", {})
        if isinstance(projects_data, dict):
            for project_type in ["professional", "personal"]:
                for project in projects_data.get(project_type, []):
                    project_text = json.dumps(project, ensure_ascii=False).lower()
                    for skill in skill_metrics:
                        if skill in project_text:
                            skill_metrics[skill]["contexts"].add("projets")
                            skill_metrics[skill]["frequency"] += project_text.count(skill)
        for formation in self.cv_data.get("formations", []):
            formation_text = json.dumps(formation, ensure_ascii=False).lower()
            for skill in skill_metrics:
                if skill in formation_text:
                    skill_metrics[skill]["contexts"].add("formations")
                    skill_metrics[skill]["frequency"] += formation_text.count(skill)
        final_scores = []
        for skill, metrics in skill_metrics.items():
            if metrics["frequency"] == 0:
                continue
            
            context_score = max((self.CONTEXT_VALUES.get(c, 0) for c in metrics["contexts"]), default=0.1)
            if len(metrics["contexts"]) > 1:
                context_score = 1.0

            normalized_frequency = self._normalize_score(metrics["frequency"])
            normalized_depth = self._normalize_score(metrics["max_duration"])

            final_score = (self.ALPHA * context_score) + \
                          (self.BETA * normalized_frequency) + \
                          (self.GAMMA * normalized_depth)

            final_scores.append({
                "skill": metrics["original_name"],
                "score": round(final_score, 2),
                "details": {
                    "context_score": round(context_score, 2),
                    "contexts_found": list(metrics["contexts"]),
                    "frequency": metrics["frequency"],
                    "max_duration_years": round(metrics["max_duration"], 1)
                }
            })

        final_scores.sort(key=lambda x: x["score"], reverse=True)
        logger.info(f"Scoring terminé pour {len(final_scores)} compétences.")
        
        return {"analyse_competences": final_scores}
OptimizedContextualScoringEngine = ContextualScoringEngine